In recent years, attention has been drawn to distributed computing aimed at efficient use of resources and the like. Such distributed computing is called cloud and offers various types of services according to such models as Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). In distributed computing, a plurality of computing devices constituting a cell execute a task instructed by a client in coordination with one another, and therefore consensus may be achieved among the computing devices. One example of algorithms for achieving consensus in this manner is Paxos described in, for instance, PLT 1 through PLT 3.
Paxos is a consensus algorithm according to which each of computing devices constituting a cell acts as a leader or an agent. Paxos works such that, before the distributed computing system executes a predetermined function, the leader checks whether or not computing devices constituting a quorum (majority) agree on the function; if consensus is achieved to that effect, the execution is started. Use of such an algorithm allows easily synchronizing the computing devices, thus achieving high availability.